An integrated circuit (IC) is an assembly of discrete devices such as resistors, transistors, capacitors, etc. As an example, the transistor is a device comprising on-off properties that act as a power switch to the IC. A commonly known transistor to those skilled in the art is a metal-oxide-semiconductor field effect transistor (MOSFET).
Generally speaking, a typical MOSFET comprises a silicon substrate with a thin layer of thermally grown oxide used to isolate the substrate from conductive electrode. The metal electrode, generally referred to as a gate electrode, controls the on-off properties of the working device. The substrate including the gate electrode further comprises an oxidized gate dielectric layer on the sidewalls of the gate electrode and low-doped drain regions formed by implanting ions into the substrate of both sides of the gate dielectric layer. A spacer of an insulating layer is generally formed adjacent to opposite sidewalls of the gate electrode and the oxidized gate dielectric layer. An ion implantation process is performed to form a source and a drain at predetermined positions of the substrate besides the at least two spacers.
A conventional spacer commonly known to those skilled in the art is a D-shaped spacer. As the semiconductor industry continues to increase the density of devices manufacturable on a semiconductor substrate, the device features continue to shrink below the quarter-micron size. The critical dimensions between the gate electrodes are decreased, making it more difficult to achieve void-free fill between the spacers adjacent to the gate electrodes. In addition, channel stress-related and silicide stress-related problems increase with decreasing dimension. It is desired to fabricate semiconductor devices using reliable processing methods including a reduction in the number of steps required to fabricate a semiconductor device, reduced defects, lower thermal budget, reduced channel stress, and reduced resistivity at the source/drain region.
The disclosed invention provides space methods of forming this spacer, and more specifically a method of forming a spacer.
It will be appreciated that for simplicity and clarity of illustration, elements illustrated in the Figures have not necessarily been drawn to scale. For example, the dimensions of some of the elements are exaggerated relative to other elements for clarity. Further, where considered appropriate, reference numerals have been repeated among the Figures to indicate corresponding or analogous elements.
FIGS. 1 through 18 depict, in accordance with various embodiments of the disclosures made herein, the formation of a semiconductor device with a self aligned spacer, and a method of forming such a self aligned spacer during manufacture of a semiconductor device. The method provides for improved thermal budget, reduced channel stress, and reduced resistivity at the source/drain region.
FIG. 1 is a cross-sectional view illustrating an exemplary embodiment of a portion of a semiconductor device 11. The device of FIG. 1 includes a substrate 10. In the assorted embodiments, substrate 10 is a semiconductor device substrate such as silicon, gallium arsenide, or other appropriate substrate, on which the illustrated features have been formed. For reference purposes, substrate 10 is illustrated as having a horizontal surface. A structure 16 is formed on the substrate 10, wherein the structure 16 comprises a sidewall portion 17 that is substantially orthogonal to the surface of the substrate 10, and a top portion 19 that is substantially parallel to the surface of the substrate 10. In a more specific embodiment, the structure 16 is a gate structure comprising a gate electrode 40 overlying a gate dielectric layer 42, which forms the sidewall portion 17. In an embodiment, the gate electrode 40 is a polysilicon conductor and the gate dielectric layer 42 is an oxide layer, such as silicon dioxide, hafnium oxide, or a hafniumsilicate compound. In another embodiment, the gate electrode 40 comprises a layer containing at least one of a heavy metal or oxygen atoms, or other suitable gate material known to those of skill in the art. By way of example, the heavy metal atoms include Ruthenium or Indium. More specifically, the gate electrode 40 can comprise Ruthenium oxide or Iridium oxide.
FIG. 2 is a cross-sectional view illustrating a liner oxide layer 21 formed over the substrate 10 and gate electrode 40 of semiconductor device 11 according to an embodiment of the present disclosure. Liner oxide layer 21 may be formed using low-pressure chemical vapor deposition (LPCVD) or plasma enhanced chemical vapor desposition (PECVD) techniques, and ranges in thickness from about 50 to 300 angstroms. In one embodiment, the thickness of liner oxide layer 21 ranges from about 60 to 200 angstroms.
FIG. 3 is a cross-sectional view illustrating a dielectric layer 22 formed over the liner oxide layer 21 of semiconductor device 11. In an embodiment, the dielectric layer 22 is a nitride layer ranging in thickness from about 150 to 500 angstroms. In an exemplary embodiment, the thickness of dielectric layer 22 ranges from about 150 to 350 angstroms, and may be formed using LPCVD or PECVD techniques. Further, precursors for the nitride deposition include silane, dichlorosilane, hexachlorosilane, or BTBAS.
FIG. 4 is a cross-sectional view illustrating formation of an additional layer 24 over the semiconductor device 11 according to at least one embodiment of the present disclosure. In one embodiment, additional layer 24 may be comprised of an oxide layer ranging in thickness from about 500 to 1000 angstroms. For example, the oxide layer can include silicon dioxide. In another embodiment, additional layer 24 may be a silicon-containing layer ranging in thickness from about 300 to 900 angstroms. For example, the silicon-containing additional layer 24 can include polysilicon or amorphous silicon. Additional layer 24 serves the purpose of being sacrificial to the degree required to shape a forming spacer, as seen in FIG. 5. FIGS. 5 through 10 reference forming spacers made from an oxide additional layer 24, while FIGS. 11, 12, and 15 through 18 reference forming spacers made from a silicon-containing additional layer 24.
FIG. 5 illustrates a cross-sectional view showing the forming spacers 25 adjacent to the sidewall portion of structure 16 including the gate electrode 40 that remain after etching away portions of additional layer 24. In one embodiment, the forming spacers 25 are oxide portions that substantially fill an interior angle space 23 defined by the horizontal and vertical portions of the dielectric layer 22. Portions 15 of the dielectric layer 22 which do not have an overlying forming spacer 25, are referred to as unprotected portions 15 of the dielectric layer 22. For this particular embodiment, a process to etch away portion of additional layer 24 to form spacers 25 is a process selective to nitride that includes using C4F8 as a main etch gas and is commonly known to those skilled in the art. In another embodiment, the forming spacers 25 are silicon-containing portions that substantially fills an interior angle space 23 defined by the horizontal and vertical position of the dielectric layer 22. For this case, the silicon-containing forming spacers 25 can be sacrificial portions that are removed after its purpose has been served. In particular, the silicon-containing portions are removed using a process that includes HBr, Cl2, and HeO2 and is also commonly known to those skilled in the art.
FIG. 6 is a cross-sectional view illustrating the semiconductor device 11, wherein at least part of the unprotected portions 15 (FIG. 5) of the dielectric layer 22 are removed to form at least one spacer 27. In one embodiment, the spacer 27 is an L-shaped spacer 27 having a sidewall portion 28 and a substrate portion 29. More specifically, the sidewall portion 28 of the spacer 27 is a vertical portion parallel to the sidewall portion 17 of structure 16 (FIG. 1) including the gate electrode 40. Further, the substrate portion 29 of the spacer 27 is a horizontal portion approximately orthogonal to the sidewall portion 17 of structure 16 including the gate electrode 40 and is substantially parallel to the original surface of the substrate 10. Portions 30 of liner oxide layer 21 that do not have an overlying L-shaped spacer 27 are referred to as unprotected portions 30 of the liner oxide layer 21. In one embodiment, the sidewall portion 28 of the L-shaped spacer 27 may be recessed as shown in FIG. 6.
By way of example, the distance between the uppermost section of the sidewall portion 28 and the upper surface of the gate electrode 40 range from about 200 to 300 angstroms. The recessed L-shaped spacer 27 enhances silicide growth in the following fabrication sequence. It should be noted that although forming spacer 25 is illustrated in FIG. 6 as extending above the uppermost section of sidewall portion 28 of L-shaped spacer 27, in other embodiments, forming spacer 25 may be on the same level as, or recessed slightly below, the uppermost section of sidewall portion 28 of L-shaped spacer 27. Final size of forming spacer 25 is dependant upon the extent of etch conducted upon it.
In the described embodiments, the etch process for removing at least part of the unprotected portions 15 of the dielectric layer 22 can utilize a dry etch process. In the assorted embodiments, the etch process involves the substrate 10 to be placed in an exemplary process chamber capable of performing a dry, anisotropic etch. The chamber can be a capacitively coupled chamber or an inductively coupled chamber. In one embodiment, the process steps of forming the forming spacer 25 by removing portions of additional layer 24 and the process steps of forming the L-shaped spacer 27 by removing at least part of the unprotected portions 15 of the dielectric layer 22 are executed in one process chamber. In another embodiment, the process steps of forming the forming spacer 25 and the L-shaped spacer 27 are performed in at least two different process chambers.
The chamber is pressurized to a value that enhances a uniform etch and the substrate 10 is adjusted to a temperature such as about 10C to 30C. In the case in which the dielectric layer 22 is a nitride layer, the resulting L-shaped spacer 27 comprises a nitride layer. A fluorine gas source and an oxygen gas source then flows a fluorine-containing gas mixture, which may optionally include an oxygen-containing gas mixture, into the pressurized chamber. The gas flow rate ratio may vary depending on the particular embodiment and the desired selectivity for a particular application.
In one embodiment, a process selective to both oxide and silicon layers includes using CH3F and O2. The gas mixture comprising CH3F and O2 flows into the chamber with a gas flow rate ratio having a value within the range of approximately 1:2 and 1:6, with a preferred range of 1:2 and 1:3. For this particular case, the chamber is pressurized to a value within the range of approximately 20 and 250 milli Torr (mTorr) and, more specifically, 60 to 100 mTorr.
In another embodiment, the use of an oxygen gas source is optional. By way of example, a process that is sufficiently selective to an oxide and silicon layer comprises CF4, and CHF3. The gas flow rate ratio of CF4 and CHF3 is a value within the range of approximately 1:0 and 0:1 and the chamber is pressurized to a value within the range of approximately 50 and 250 mTorr. Alternatively, a process selective to an oxide layer but not a silicon layer includes using CF4 and HBr. For this case, the substrate 10 is adjusted to a temperature such as about 50C to 80C. The gas flow rate ratio of CF4 and HBr is a value within the range of approximately 5:1 and 2:1 and the chamber is pressurized to a value within the range of approximately 50 and 150 mTorr.
After a sufficient time for the pressure and gas flow to stabilize, an excitation power is applied to sustain plasma for a sufficient period of time needed to etch away the unprotected area of a layer, such as the unprotected portion 15 of the dielectric layer 22 shown in FIG. 5. In one embodiment, an RF source power generator (not shown) coupled to an inductive antenna coils provides power to generate reactive species and controls the plasma density, which, in turn, enhances chemical reactions. An RF bias power generator controls the ion bombardment of the surface of device 11, and enhances a more physical etch. In another embodiment, only a single RF power generator is used and the etching mechanism may be either reactive or physical.
The period of time in which the excitation power sustains plasma needed to etch away the unprotected area of the dielectric layer 22 varies depending on the thickness of the dielectric layer 22. By way of example, the dielectric layer 22 is a nitride layer. More specifically, the dielectric layer 22 is a nitride layer having a thickness within the range of 150 and 500 angstroms. For this particular case, the etch rate is a value within the range of 300 and 900 angstroms per minute and, therefore, the excitation power is sustained at a value within the range of approximately 10 and 100 seconds.
In one embodiment, unprotected portions 30 of liner oxide layer 21 may be etched such that only the protected portion 20 remains, as seen in FIG. 7. FIG. 7 is a cross-sectional view of the portion of a transistor of semiconductor device 11 after etching away the unprotected portions 30 of the liner oxide 21. The protected portion 20 of liner oxide layer 21 that remains is also L-shaped.
In the various embodiments, a metal layer such as one of titanium, cobalt, nickel, or similar group metals may be formed over the device 11 of FIG. 8, including over the forming spacer 25, L-shaped spacer 27, remaining portion of liner oxide 20, and the top of gate electrode 40. Annealing the device 11 of FIG. 8 at an elevated temperature causes the metal layer to react with the underlying surfaces of substrate 10 and the polysilicon gate electrode 40 during a silicidation process. More specifically, the heat reaction with the metal layer and the silicon present in substrate 10 and gate electrode 40, form silicide layers 92, illustrated by cross-section in FIG. 8. It should be noted that the remaining metal layer, which does not react at elevated temperature, is removed and is thus not illustrated in FIG. 8.
FIG. 9 illustrates a cross-sectional view of device 11 further including the addition of a pre-metal dielectric layer 108. In one embodiment, the pre-metal dielectric layer 108 includes at least one of an oxide layer, a boron silicon glass (BSG) layer, a boron phosphorous silicon glass (BPSG) layer or a phosphorous silicon glass (PSG) layer. Pre-metal dielectric layer 108 is formed over the device 11 of FIG. 8 to accommodate a subsequent metal layer (not illustrated). It should be noted that device 11 of FIG. 9 is a transistor, however, the formation of source/drain regions is not illustrated in FIG. 9.
In an alternate embodiment of the present disclosure, the pre-metal dielectric layer 108 comprises a composite layer 111 including a first layer containing nitrogen 109 and a second layer 110, as illustrated in the cross-sectional view of FIG. 10. In particular, the first layer 109 of the composite layer 111 is between the forming spacer 25 overlying the spacer 27 and the second layer 110. By way of example, the first layer 109 is a silicon nitride layer or a siliconoxynitride layer. The second layer 110 comprises at least one of an oxide layer, a boron silicon glass (BSG) layer, a boron phosphorous silicon glass (BPSG) layer or a phosphorous silicon glass (PSG) layer.
FIG. 11 is a cross-section view illustrating a partially formed transistor of semiconductor device 11 according to an embodiment of the disclosure. In the illustrated embodiment of FIG. 11, the various processing steps up to the formation of the L-shaped spacer 27 and remaining portion of liner oxide 20 have been completed, as was presented in FIGS. 4 through 7. Additionally, in FIG. 11, removal of the forming spacer (item 25 in FIGS. 5 through 7) which substantially filled an interior angle space defined by the horizontal and vertical position of the spacer 27 has been accomplished. For this particular embodiment, the forming spacer was a silicon-containing forming spacer. An exemplary technique for removing the (silicon or oxide) forming spacer includes reactive ion etching (RIE) dry etching.
FIG. 12 illustrates a cross-sectional view of the semiconductor device of FIG. 11 having a silicide and pre-metal dielectric layer formed in a similar manner as discussed with reference to FIGS. 8 and 9. Alternately, the pre-metal dielectric layer may comprise a composite layer such as was discussed with reference to FIG. 10, including a first layer containing nitrogen, and a second layer.
Referring now to FIG. 13, a cross-sectional view of a semiconductor device according to at least one embodiment of the present invention is presented. In this particular embodiment, semiconductor device 111 comprises two partially formed transistors 103 and 104. The distance of separation between the gate sidewalls of partially formed transistors 103 and 104 is denoted by the letter d, and in one embodiment is less than 150 nanometers, with a preferred distance being at or less that 120 nanometers. Device 111 comprises a substrate 110, a first gate structure 215 and a second gate structure 216 formed on the substrate 110, at least one gate dielectric layer 220, a liner oxide layer 121 and a spacer dielectric layer 122 which has been formed overlying liner oxide layer 121. Substrate 110 may be active silicon or field isolation material. There is no oxide or silicon-containing forming spacer utilized in this particular embodiment.
The spacer dielectric layer 122 is selectively etched to form an L-shaped spacer 127 as shown in the cross-sectional illustration of FIG. 14. In one embodiment, spacer dielectric layer 122 is a nitride layer. More specifically, the spacer dielectric layer 122 can be a nitride layer having a thickness within the range of approximately 150 and 500 angstroms. Similar to the dielectric layer 22 of FIG. 5, the spacer dielectric layer 122 can be formed using an PECVD or LPCVD technique. The underlying liner oxide 121, when used, would generally have a thickness in the range of approximately 20 angstroms to 200 angstroms.
For this particular embodiment, the process for forming the L-shaped spacer 127 from the dielectric layer 122 can utilize a dry etch process. The etch process involves the substrate 110 to be placed in an exemplary chamber capable of performing a dry, anisotropic etch. For example, the chamber can be a capacitively coupled chamber or an inductively coupled chamber. The chamber is pressurized to a value that enhances a uniform etch and the substrate 110 is adjusted to a temperature such as about 10C to 30C.
A fluorine gas source and an oxygen gas source then flows a fluorine-containing gas mixture and an oxygen-containing gas mixture into the pressurized chamber. The gas flow rate ratio may vary depending on the particular embodiment and the desired selectivity for a particular application. By way of example, the gas mixture comprises CH3F and O2. The CH3F and O2 gas mixture flows into the chamber with a gas flow rate ratio (CH3F: O2) having a value within the range of approximately 2:1 and 5:1, with a preferred range of 3:1 and 4:1. For this particular case, the chamber is pressurized to a value within the range of approximately 20 and 500 mTorr and, more specifically, 60 to 100 mtorr.
In the described embodiment, the CH3F-rich gas mixture allows the L-shaped spacer 127 to be formed without employing the use of a sacrificial forming spacer. Variations and different embodiments of the L-shaped spacer 27 of FIG. 9 are applicable in the L-shaped spacer 127. In addition, the resulting benefits from the L-shaped spacer 27 of FIG. 9 are applicable to the L-shaped spacer 127. Examples of resulting benefits include a reduction in the number of steps required to fabricate a semiconductor device, reduced defects, lower thermal budget, reduced channel stress, and reduced resistivity at the source/drain region. The process used to form spacer 127 can reduce the number of steps used in the formation of the L-shaped spacers. The resulting spacer 127 has a bulging, which, with respect to the horizontal portion 128 of the spacer 127, results in a profile which varies gradually in thickness from a maximum thickness immediately adjacent the vertical portion of the L-shaped spacer to a portion of the L-shaped spacer furthest from the vertical-portion of the L-shaped spacer. The resulting profile of the horizontal portion of the spacer 127 varies gradually to provide for an average thickness of that is 50 to 85 percent of the maximum thickness, and, more specifically, 60 to 75 percent of the maximum thickness. In addition, one embodiment results in the formation of the horizontal portion 128 having a length that is between 80 and 150 percent of the deposition thickness of the layer 122 from which the spacers are formed.
FIG. 15 illustrates a cross-sectional view of a partial transistor 1100 having a lightly doped source/drain region 230 (also referred to as a source/drain extension) in a substrate 101. Partial transistor 1100 is formed in the same manner as the partial transistor 11 illustrated in FIG. 11. Although not illustrated in FIG. 15, it is possible to have various halo implants underlying doped regions 230 in order to control short channel effects.
FIG. 16 is a cross-sectional view illustrating the formation deeply formed by doping source/drain regions 660 within portions of the substrate 101 adjacent to the spacers 270. Doped regions 660 are created by ion implantation 305 and is commonly known to those skilled in the art.
In FIG. 17, substrate 101 is doped by ion implantation 305 to form intermediate doped regions 760 and deeply formed source/drain regions 660. In an embodiment, the intermediate doped regions 760 are intermediate source/drain regions formed by doping the substrate 101, wherein the dopants are implanted through the substrate portion of the spacer 270. The formation of intermediate doped regions 760 is a function of both the thickness of the lower portion of the L-shaped spacer and the energy of the implanted ions (implant power). By way of example, the substrate portion of the L-shaped spacer 270 comprise a thickness within the range of approximately 150 to 500 angstroms and the underlying liner oxide layer 200 includes a thickness within the range of approximately 50 and 300 angstroms. For this assorted embodiment, possible implant species for a PMOS are Boron (B), Indium (In) and Boron diflouride (BF2). In particular, to implant Boron with an ion dose within the range of 1e13/cm2 and 1e15/cm2, the ion energy applied is a value within the range of about 5 keV and 15 keV. Possible implant species for an NMOS are Arsenic (As), Phosphorous (Ph) and Antimony (Sb). Similarly, to implant Arsenic with an ion dose within the range of 1e13/cm2 and 1e15/cm2, the ion energy applied is a value within the range of about 10 keV and 100 keV.
The intermediate doped regions 760 is not formed as deeply into the substrate 101 as the deeply formed source/drain regions 660, since the L-shaped spacer resides over the intermediate doped regions 760.
In the described embodiments, the depth of the intermediate doped regions 760 can be a value within the range of approximately 50 angstroms to 500 angstroms, while the source/drain region 660 will have a greater depth, generally in the range of 300 to 1000 Angstroms or greater. In another embodiment in which the ion energies are at the lower end of its range and/or the thickness of the oxide and nitride layers are at the upper end of its range, the ion does not penetrate through the substrate portion of the spacer 270 to the silicon. In particular, the deeply formed source/drain regions 660 have an average depth that is greater than the average depth of the intermediate source/drain region 760. Likewise, the intermediate source/drain region 760 has an average depth that is greater than the source/drain extension region 230.
FIG. 18 is a cross-sectional view of a substantially completed transistor 1100 according to an embodiment of the present disclosure. A pre-metal dielectric layer 1108 is applied over the partially formed transistor device 1100 of FIG. 17. In one embodiment, the pre-metal dielectric layer 1108 may be comprised of a composite layer (not illustrated). A metal structure 990 has been formed over a portion of the pre-metal dielectric layer 1108. Formation of metal structures, such as metal structure 990, can be performed using any appropriate process as would be readily known in the art. While metal structure 990 is illustrated as an inlaid structure, it will be appreciated that non-inlaid structures may be used as well.
In the foregoing specification, the invention has been described with reference to specific embodiments. However, one of ordinary skill in the art appreciates that various modifications and changes can be made without departing from the scope of the present invention as set forth in the claims below. Accordingly, the specification and figures are to be regarded in an illustrative rather than a restrictive sense, and all such modifications are intended to be included within the scope of present invention.
Benefits, other advantages, and solutions to problems have been described above with regard to specific embodiments. However, the benefits, advantages, solutions to problems, and any element(s) that may cause any benefit, advantage, or solution to occur or become more pronounced are not to be construed as a critical, required, or essential feature or element of any or all the claims.